AI forecasting secret: trendlines are so steep that a 10x error shifts timelines by only a year
harris_edouard · x · 2026-10-24
Edouard Harris points out a counterintuitive property of AI forecasting: exponential trendlines are so violent that being off by a factor of 10 on the growth rate shifts your timeline forecast by only a year or two — making timeline predictions strikingly insensitive to large errors in capability estimates, and helping explain why forecasters' timelines converge.
More from AGI Musings
- Mathematician Danny Calegari: LLMs Mirror You Back, But Behave Better — fortnow · 2026-10-24
- No One Is Impressed by Good Results: Why AI-Proved Theorems Teach You Nothing — fortnow · 2026-10-24
- If Claude Had the Idea, Say So 'For Your Own Personal Soul': AI and Math Ethics — fortnow · 2026-10-24
- Top Mathematician Shows How He Uses Claude to Do Research and Build Talks — fortnow · 2026-10-24
- UChicago mathematician Danny Calegari shows how he uses Claude for math collaboration and slide generation — fortnow · 2026-10-24
- New paper introduces "bottleneck pressure" to measure where work AI can't do gains value — soumitrashukla9 · 2026-10-24